Opioid substitution therapy (OST) is the most effective treatment option for opioid addiction. Medicaid beneficiaries have high rates of opioid addiction and Medicaid programs are among the largest funders of OST. Buprenorphine has recently been introduced as an effective alternative for OST and has been shown to expand treatment access and participation, but it is expensive, and Medicaid programs in many states have begun to restrict access to buprenorphine treatment without knowing how those restrictions affect patient outcomes or overall state spending. To date, there have been no studies of buprenorphine's impact on Medicaid expenditures, relapse related service use or criminal justice system encounters. The proposed study has two primary aims: (1) to study the impact of a 2008 Medicaid policy change in Massachusetts that introduced a prior authorization requirement related to buprenorphine dose levels, and (2) to compare costs and outcomes of Medicaid funded treatments for opioid dependence, including buprenorphine, methadone maintenance and outpatient drug free treatment as well as no treatment. Outcomes of interest for Aim 1 include dose levels, switching to other forms of treatment, use of relapse-related services (inpatient, emergency room and detoxification), Medicaid expenditures and criminal justice encounters. Outcomes of interest for Aim 2 include relapse-related service use, Medicaid expenditures, state expenditures, arrests and incarceration. Combining Medicaid claims and other administrative data on publicly funded treatment with arrest and incarceration data from 2003 through 2009, we propose to use generalized estimating equations (GEE) and individual growth models to measure changes in outcome variables after adoption of the 2008 Medicaid policy, controlling for a range of patient and policy factors. Comparative analysis of treatment effects and expenditures will use GEE and instrumental variables to measure differences between the treatments within an intent-to-treat format similar to that used in randomized clinical trials. The study will include approximately 45,000 Massachusetts Medicaid beneficiaries diagnosed with opioid dependence during the study timeframe and will be the largest study of Medicaid-funded OST to date. Study findings will be useful for state and Medicaid policy makers, administrators and providers across the US who often must make coverage or treatment decisions in an environment where pressure to contain spending is intense and information on costs and benefits is lacking. PUBLIC HEALTH RELEVANCE: Medicaid beneficiaries have higher rates of addiction than other insured groups, and Medicaid programs are the largest funders of opioid substitution therapy (OST). Buprenorphine, which was approved for OST by the Federal Drug Administration in 2002, has not been studied longitudinally in a Medicaid population. The proposed study will be the first to measure buprenorphine's impact on Medicaid expenditures, use of relapse related services, and criminal justice involvement. It will also be the first to measure the effect of prior authorization policies that states are increasingly imposing on buprenorphine. New knowledge gained from the study will be useful for developing more cost-effective policies and practices for managing opioid substitution treatment.